This project examines the role of patient factors and treatment alliance (TA) in the outcome of depression in primary care. The overarching hypothesis is that patients' attitudes about depression, help-seeking, and antidepressant medication, and the quality of their social support influence the TA, treatment adherence, and ultimately the outcome of care for depression. The project involves a mixed method triangulation design in which patients newly identified as depressed are followed for a year with either a quantitative or qualitative methodology. Participating physicians will record all patients they detect as depressed. Following the index visit, patients will be given an initial telephone screening for eligibility. Patients who are deemed incident cases, i.e., who have not received treatment for depression in the previous 6 months, will be scheduled for a comprehensive telephone interview within 2 weeks. A smaller number of subjects will be selected for the qualitative component of the study via purposeful sampling. For the patients assigned to the quantitative arm of the study, this interview will include modules of the SCID and a standardized assessment of attitudes and experiences relevant to depression and its treatment. They will also be provided with a mailback questionnaire and will receive brief assessments at 6 weeks, and 3, 6, and 9 months. A final comprehensive interview and questionnaire at 12 months will reassess symptoms and diagnosis; attitudes, beliefs, and social support; the intervening experience of depression, its treatment and the TA. Relevant physician attitudes and treatment practices will be assessed before patient recruitment commences and at the completion of the study. Patients selected for the qualitative component will participate in unstructured interviews concerning the index encounter, relevant past experiences, and views of their physician, depression and the treatment being offered them, with reassessment at 6 and 12 months. At the end of the 12-month interview clinical diagnosis will be ascertained. Qualitative data analysis will be based on grounded theory methodology. This theoretical model will be compared (triangulated) with the longitudinal structural model developed in the quantitative phase of study. In this way, both quantitative and qualitative results and interpretations will inform an understanding of patient depression, adherence, and therapeutic alliance. Results of the project will highlight the need for continued efforts to change lay beliefs concerning the nature and treatment of depression. They will also demonstrate that primary care may not be the optimal setting for treating all patients, and explain why this is so. For the many patients who are treated in primary care, the project will contribute to the development of more collaborative, and, thus, more effective disease management by specifying the themes and barriers that should be addressed in tailored messaging, education, and other patient-oriented treatment strategies.